Activity

Automatic sleep stage detection can be performed using a variety of input signals from a polysomnographic (PSG) recording. In this study, we investigate the effect of different input signals on the performance of feature-based automatic sleep stage classification algorithms with both a Random Forest (RF) and Multilayer Perceptron (MLP) classifier. Combinations of the EEG (electroencephalographic) signal and ECG (electrocardiographic), EMG (electromyographic) and respiratory signals as input are investigated as input with respect to using single channel and multi-channel EEG as input. The Physionet “You Snooze, You Win” dataset is used for the study. The RF classifier consistently outperforms our MLP implementation in all cases and is positively affected by specific signal combinations. The overall classification performance using a single channel EEG is high (an accuracy, precision and recall of 86.91 %, 89.52%, 86.91% respectively) using RF. The results are comparable to the performance obtained using six EEG channels as input. Adding respiratory signals to the inputs processed by RF increases the N2 stage detection performance with 20%, while adding the EMG signal improves the accuracy of the REM stage detection with 5%. Our analysis shows that adding specific signals as input to RF improves the accuracy of specific sleep stages and increases the overall performance. Using a combination of EEG and respiratory signals we achieved an accuracy of 93% for the RF classifier.We interface the head modelling, coil models, Graphical User Interface (GUI), and post-processing capabilities of the SimNIBS package with the boundary element fast multipole method (BEM-FMM), implemented in a MATLAB-based module. The resulting pipeline combines the best of both worlds the individualized head modelling and ease-of-use of SimNIBS with the numerical accuracy of BEM-FMM. The corresponding TMS (transcranial magnetic stimulation) modeling package is developed and made available online. It imports a SimNIBS surface segmentation and a coil field, and then exports electric-field values in selected surfaces or volumes. Additional information is also made available, such as discontinuous compartment surface electric fields and associated surface electric charge distributions.In addition to socioeconomic influences, biological factors are believed to play a role in health disparities. In this paper, we investigate miRNA, mRNA, and DNA methylation patterns that contribute to disparities in hepatocellular carcinoma (HCC). This is accomplished by integration of mRNA-Seq, miRNA-Seq, and DNA methylation data we acquired by analysis of liver tissues from 30 HCC patients consisting of European Americans (EAs), African Americans (AAs), and Asian Americans (Asians). Mixed-ANOVA models are applied to identify miRNAs, mRNAs, and DNA methylation sites that are significantly altered in tumor vs. adjacent normal tissues in a race-specific manner. Through integrated analysis, a refined list of differentially expressed mRNAs is obtained by selecting those that are targets of differentially expressed miRNAs and consist of promoter regions that are differentially methylated.Identifying differentially expressed subpathways connected to the emergence of a disease that can be considered as candidates for pharmacological intervention, with minimal off-target effects, is a daunting task. In this direction, we present a bilevel subpathway analysis method to identify differentially expressed subpathways that are connected with an experimental condition, while taking into account potential crosstalks between subpathways which arise due to their connectivity in a combined multi-pathway network. The efficacy of the method is demonstrated on a hematopoietic stem cell aging dataset, with findings corroborated using recent literature.With cancer being one of the main remaining challenges of modern medicine, a lot of effort is put towards oncology research. Since early diagnosis is a highly important factor for the treatment of many types of cancer, screening tests have become a popular research subject. Technical and technological advances have brought down the price of genome sequencing and have led to an increase in understanding the relationship between DNA, RNA and tumor sites. D-Galactose cost These advances have sparked an interest in personalized and precision medicine research. In this work, we propose a deep neural network classifier to identify the anatomical site of a tumor. Using 27 TCGA miRNA stem-loops cohorts, we classify tumors in 20 anatomical sites with a 96.9% accuracy. Our results demonstrate the possibility of using stem-loop expression data for accurate cancer localization.In this paper, we introduce a new dataset for cancer research containing somatic mutation states of 536 genes of the Cancer Gene Census (CGC). We used somatic mutation information from the Cancer Genome Atlas (TCGA) projects to create this dataset. As preliminary investigations, we employed machine learning techniques, including k-Nearest Neighbors, Decision Tree, Random Forest, and Artificial Neural Networks (ANNs) to evaluate the potential of these somatic mutations for classification of cancer types. We compared our models on accuracy, precision, recall, and F1-score. We observed that ANNs outperformed the other models with F1-score of 0.36 and overall classification accuracy of 40%, and precision ranging from 12% to 92% for different cancer types. The 40% accuracy is significantly higher than random guessing which would have resulted in 3% overall classification accuracy. Although the model has relatively low overall accuracy, it has an average classification specificity of 98%. The ANN achieved high precision scores (> 0.7) for 5 of the 33 cancer types. The introduced dataset can be used for research on TCGA data, such as survival analysis, histopathology image analysis and content-based image retrieval. The dataset is available online for download https//kimialab.uwaterloo.ca/kimia/.Integration of multi-omics and pharmacological data can help researchers understand the impact of drugs on dynamic biological systems. Network-based approaches to such integration explore the interaction of different cellular components and drugs. However, with ever-increasing amounts of data, processing these high-dimensional biological networks requires powerful tools. We investigate whether network embeddings can address this problem by providing an effective method for dimensionality reduction in drug-related networks. A neural network-based embedding method is employed to encode protein-protein, protein-disease, drug-drug and drug-disease networks for the prediction of novel drug-target interactions. We found that drug-target interaction prediction using embeddings of heterogeneous networks as input features performs comparably to state-of-the-art methods, exhibiting an area under the ROC curve of 84%, outperforming methods such as BLM-NII and NetLapRLS, and coming very close to the best performing network methods such as HNM, CMF and DTINet.